Aggregate stressors to 10 km Mollweide

library(raster)
source('https://raw.githubusercontent.com/oharac/src/master/R/common.R')

source(here('common_fxns.R'))

dir_str_mol <- '/home/shares/ohi/git-annex/impact_acceleration/stressors'
dir_str_10km  <- file.path(dir_bd_anx, 'layers/stressors_10km')

1 Summary

Collect stressor layers from CHI 2019 and reproject/aggregate them to the native resolution of the species range maps. This can be a simple aggregation 11x as is done for the ocean, EEZ, MEOW, etc rasters.

1.1 Loop over all stressor layers, aggregate to spp CRS

Read in each raster, aggregate to a raster at the species projection and resolution. Because the projection is Mollweide for the source and target, and the base raster for species maps was created the same way, we can simply aggregate the original Mollweide CHI maps up by the same factor. For spp maps, the aggregation factor was 11\(\times\) to approximate 100 km² cells: \(.934 \times 11 = 10.28 \text{ km}; \; (.934 \times 11)^2 = 105.6633 \text{ km}^2\).

Note: here we are aggregating using mean value.

1.2 Clip Sea Level Rise stressors to shallow waters only

SLR stressor map covers the whole ocean; but impacts would only affect coastal species. Since several species sensitive to SLR are wide-ranging (e.g. seabirds, turtles) we don’t want to suggest SLR impacts far at sea, so we will clip the SLR stressor layers to shallow waters only.

1.3 Mask SST rasters to exclude sea ice

SST data is problematic in the presence of sea ice, not least because the ocean water is not on the surface!

1.4 Map all stressors in gifs